The Market Approach to Valuing Businesses (Second Edition)

BV: Case Analysis Completed Transaction & Guideline Public Comparable MARKET APPROACH The Market Approach to Valuing Businesses (Second Edition) Shannon P. Pratt This material is reproduced from The Market Approach to Valuing Businesses (Second Edition) by Shannon P. Pratt with permission of John Wiley & Sons, Inc. 1995 2013 by National Association of Certified Valuators and Analysts (NACVA). All rights reserved. Market Chapters 1 Used by Institute of Business Appraisers with permission of NACVA for limited purpose of collaborative training. 2013.v2

MARKET APPROACH BV: Case Analysis Completed Transaction & Guideline Public Comparable 2 Market Chapters 1995 2013 by National Association of Certified Valuators and Analysts (NACVA). All rights reserved. 2013.v2 Used by Institute of Business Appraisers with permission of NACVA for limited purpose of collaborative training.

Selecting, Weighting, and Adjusting Market Value Multiples Chapter 10 Size and Nature of Company Equity Value Multiples Invested Capital Multiples Availability of Data Dispersion of Market Value Multiples The Harmonic Mean as a Measure of Central Tendency Adjusting from Observed Market Value Multiples Illustrative Estimate of Value Using Mathematical Weighting Summary Many analysts tend to lean toward market value of invested capital (MVIC) multiples when valuing controlling interests and equity value multiples when valuing minority interests, although that is not always the case. As noted in Chapter 9, many analysts use only the invested capital procedure whether valuing a minority or a controlling interest. Apart from the question of control or minority value, the choice of which multiples ultimately to rely on is largely a function of two factors: 1. Size and nature of the company or industry 2. Availability and nature of comparative pricing data SIZE AND NATURE OF COMPANY Equity Value Multiples Generally speaking, when choosing among possible equity value multiples, the smaller the company, the more we choose multiples higher, rather than farther down, on the income statement. 137

138 The Market Approach to Valuing Businesses For many very small companies, especially service companies, price/sales gets the primary weight. It is usually the most reliable number for small companies, and the most available. Many buyers feel that for a given book of business, they already know how much they can bring down to the bottom line. Probably some of the best examples are property and casualty insurance agencies, which tend to have a high degree of customer persistence. The next most commonly used multiple for very small businesses is discretionary earnings (earnings before interest, taxes, noncash charges, and all compensation and benefits to one owner/operator). For many very small businesses, any income number beyond this may be negative or so small as to be meaningless. As company size begins to increase, however, discretionary earnings tend to become meaningless. Although there are no precise guidelines as to the size level at which discretionary earnings becomes meaningless, most analysts would not use it for companies valued over $5 million, and some would use it only for companies valued under $1 or $2 million. Also, the number of owner/operators is important. The definition of discretionary earnings is before compensation to one owner. To the extent that there are two, three, or more owners/operators, it becomes less meaningful. As company size continues to increase, there may be positive gross cash flow, even though earnings may be meaningless or nonexistent. This is often the case because many companies are able to bonus out most of their earnings without exceeding the reasonable compensation levels of the Internal Revenue Service. With larger companies, net income tends to become a meaningful measure of earnings. Invested Capital Multiples For invested capital multiples, similar principles tend to be true. For very small companies, there often is no debt, so the equity multiples are equal to the invested capital multiples. In other words, if there is no debt, invested capital is equal to equity. Therefore, MVIC/sales is equal to price/sales, and MVIC/discretionary earnings is equal to price/discretionary earnings. Invested capital multiples are useful when the subject company is highly leveraged or when the subject company leverage differs considerably from the guideline companies leverage. There is some tendency to prefer MVIC/EBITDA to MVIC/EBIT. This is because MVIC/EBITDA eliminates the effect of different policies with respect to accounting for noncash charges, and it is almost impossible to completely eliminate these differences by financial statement adjustments. However, both measures convey information about the companies. Also, EBIT may be available when EBITDA is not. The analyst usually will consider both measures and emphasize whatever is judged to be most meaningful.

Selecting, Weighting, and Adjusting Market Value Multiples 139 AVAILABILITY OF DATA Availability of data can be a compelling factor in choice of multiples. For many very small companies, sales or sales and discretionary earnings are the only operating performance variables available. For industries that have tended to lose money, there may be no or only a few companies that have positive net income.the preference is to use variables for which the most guideline companies have meaningful numbers. For some companies, price to gross profit (sales less cost of goods sold) provides a meaningful valuation metric. This is often positive even when EBITDA is negative. Research in 2004 by Franz Ross found that this metric had the lowest dispersion (as measured by the coefficient of variation) in many industries. 1 As a consequence, Pratt s Stats has now added this valuation multiple to its transaction database. DISPERSION OF MARKET VALUE MULTIPLES When market value multiples among companies in an industry are tightly clustered, this suggests that these are the multiples that the market pays most attention to in pricing companies and stocks in that industry. That is, the denominator that creates the tightly clustered multiple is the financial variable that tends to drive the market value. Therefore, a tight clustering of market value multiples may suggest that those multiples tend to deserve more weight than other multiples. We measure the degree of dispersion by a statistic called the coefficient of variation (CV), which is defined as the standard deviation divided by the mean. The standard deviation is the square root of the variance, and the variance is the sum of the squares of the deviations from the average. Although the formula for standard deviation is a bit complicated, most pocket calculators are programmed to compute it simply by entering the observations and pressing STD DEV. The mean, of course, is just the sum of the values of the observations divided by the number of observations. Then the CV is just a simple division. Based on the theory that the multiples with the least dispersion should get most weight, those with the lowest coefficients of variation would be accorded the greatest weight. As a practical matter, many analysts do not go through the mechanics of computing a coefficient of variation, but simply apply the principle by eyeballing the relative dispersion and emphasizing the multiples with the tightest clustering. THE HARMONIC MEAN AS A MEASURE OF CENTRAL TENDENCY As an alternative to the mean or median, the harmonic mean can be used to give equal weight to each guideline company in summarizing ratios that have stock price or MVIC in the numerator. It is the reciprocal of the average of the

140 The Market Approach to Valuing Businesses reciprocals of the guideline company multiples. Consider the situation with P/E (price/earnings) multiples of 15 and 5. The reciprocal of the P/E multiple of 15 is.0667, the reciprocal of the P/E multiple of 5 is.200, the average of the two reciprocals is.1334, and the reciprocal of the average is 7.5. This P/E multiple of 7.5 is the same as the P/E of a $200 portfolio with $100 invested (for $6.67 of earnings at a P/E multiple of 15) in the first company, and $100 invested (for $20 of earnings at a P/E multiple of 15) in the second company. With the $200 invested equally in each guideline company, the total earnings are $26.67, and the P/E multiple is 7.5. Although the harmonic mean is not used frequently, probably because it is unfamiliar to most readers of valuation reports, it is conceptually a very attractive alternative measure of central tendency. ADJUSTING FROM OBSERVED MARKET VALUE MULTIPLES In Chapter 9, we left our market value tables showing means, medians, standard deviations, and coefficients of variation for each market value multiple that we considered. In the earlier sections of this chapter, we have discussed the selection of relative weighting of various market multiples. We now must come to grips with making the challenging judgment as to the value to be used for each market multiple selected. As a measure of central tendency for most statistics that we use in business valuation, we tend to prefer the median over the mean, primarily to avoid distortions from outliers. We might also use the harmonic mean to avoid distortions. However, simply applying the chosen measure of central tendency of a group of guideline company multiples more often than not fails to capture differences in characteristics between our subject company and the guideline companies as a group. This is the point at which we want to use our comparative financial analysis for guidance. Each multiple used should be accorded individual attention. It is possible that some selected multiple should be above the guideline averages, while some other selected multiple should be below guideline averages. For example, a company with an above-average return on sales usually would be accorded an above-average price/sales or MVIC/sales multiple. The same company could have a below-average return on book value, which may suggest a below-average price/book value or MVIC/book value multiple. One must keep in mind that the two factors that influence the selection of multiples of operating variables the most are the growth prospects of the subject company relative to the guideline companies and the risk of the subject company relative to the guideline companies. The analyst should review the comparative financial analysis to try to assess these relationships. Two procedures are often employed in selecting multiples to apply to the subject company relative to multiples observed from the guideline companies:

Selecting, Weighting, and Adjusting Market Value Multiples 141 1. Select a subset of the guideline companies with financial characteristics (growth, margins, volatility, etc.) most like the subject company and select multiples close to those of the most comparable companies. 2. Assess the growth and risk characteristics relative to the guideline company group as a whole, and accordingly apply median multiples where appropriate or adjust multiples upward or downward from the medians based on the comparative analysis. One might adjust an observed multiple upward or downward by a percentage, or may go toward the upper or lower end of the range (e.g., use the upper quartile of the range). In extreme situations, a multiple outside the observed range may be selected. In any case, the analyst should provide a narrative explanation for the value of the multiple to be applied to the subject company s financial data. ILLUSTRATIVE ESTIMATE OF VALUE USING MATHEMATICAL WEIGHTING A simple table supporting a mathematically weighted estimate of value by the market approach is presented as Exhibit 10.1. This exhibit derives median market multiples from Exhibit 9.10 on page 133 and the fundamentals to which the multiples are applied from Exhibits 9.1 and 9.2 on page 126. We elected to use the market value of invested capital (MVIC) procedure for this illustration because the subject company (ClearSkies) is fairly highly leveraged, at least on a book-value basis: $7,500,000 book value of debt versus $5,000,000 book value of equity. Also, we judged ClearSkies as being approximately equivalent to Exhibit 10.1 Illustrative Mathematical Weighting of Indications of Value from the Transaction Method Fundamentals Median Weighted of Value for Valuation Indication of Indication of ClearSkies a Multiple b Value Weight c Value MVIC/Sales $48,000,000.50 = $24,000,000.50 = $12,000,000 MVIC/EBITDA 5,800,000 4.89 = 28,362,000.30 = 8,508,600 MVIC/BVIC 12,600,000 1.61 = 20,286,000.20 = 4,057,200 Total weighted indications of value for MVIC $24,565,800 Less: long-term debt 7,500,000 Indicated Value of Equity by Transaction Method $17,065,800 a From Exhibit 9.2. b From Exhibit 9.10. c See text for explanation.

142 The Market Approach to Valuing Businesses the average of the guideline companies in terms of growth prospects and risk, so we selected median multiples, although comparative analysis would tend to lend to different selection of multiples in most cases, as discussed in the previous section. We accorded 50% of the weight to MVIC/sales because that multiple had by far the lowest coefficient of variation, as shown in Exhibit 9.10. We accorded 30% of the weight of MVIC/EBITDA because this metric is relied on more heavily than MVIC/EBIT and the coefficients of variation were quite close between the two measures. We accorded 20% of the weight to MVIC/book value of invested capital because we felt that the assets should be given some weight. These calculations result in the $17,065,800 estimate of the market value for ClearSkies by the transaction method, as shown in Exhibit 10.1. This table is presented for illustrative purposes only, and much more analysis would be in order to make the final selection of which market multiples should be used, adjustments (if any) to the median multiples, and the relative weight to be accorded each. SUMMARY With the market approach encompassing a wide variety of market value multiples to choose from, the analyst must make a reasoned choice of which ones to use in each case. With income-related variables, the smaller the company, the more the analyst tends to rely on variables near the top of the earnings measurement spectrum, such as sales and discretionary earnings. As the company size becomes larger, the analyst tends to rely more on measures further down the income statement. For invested capital multiples, the more the accounting treatment of noncash charges varies among companies, the more the analyst tends to prefer EBITDA over EBIT, assuming that the information is available for both. The less the dispersion of observed multiples in an industry, the more the industry seems to rely on that particular multiple for pricing companies and stocks. Therefore, when using the market approach, the analyst often computes the coefficient of variation for each market value multiple, giving more weight to those multiples with the lowest coefficients of variation, or applying the general principle by eyeballing the distributions and according more weight to those that are most clustered. Finally, a value for each multiple must be selected. Here the analyst draws on the comparative financial analysis to guide the final judgment as to the appropriate value for each multiple. Notes 1. Franz Ross, Just One Thing : The Most Reliable Variable for Use in the Market Approach, Shannon Pratt s Business Valuation Update (Portland, OR: Business Valuation Resoures, 2004). Full text available at www.bvlibrary.com.